iofu728 commited on
Commit
4e392eb
1 Parent(s): 2c481ab

Feature(LLMLingua-2): update LLMLingua-2 link

Browse files
Files changed (1) hide show
  1. README.md +20 -13
README.md CHANGED
@@ -4,30 +4,30 @@ license: cc-by-nc-sa-4.0
4
 
5
  # LLMLingua-2-Bert-base-Multilingual-Cased-MeetingBank
6
 
7
- This model was introduced in the paper [**LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression** (Pan et al, 2024)](). It is a [XLM-RoBERTa (large-sized model)](https://huggingface.co/FacebookAI/xlm-roberta-large) finetuned to perform token classification for task agnostic prompt compression. The probability $p_{preserve}$ of each token $x_i$ is used as the metric for compression. This model is trained on [an extractive text compression dataset]() constructed with the methodology proposed in the [LLMLingua-2](), using training examples from [MeetingBank (Hu et al, 2023)](https://meetingbank.github.io/) as the seed data.
8
 
9
- For more details, please check the home page of [LLMLingua-2]() and [LLMLingua Series](https://llmlingua.com/).
10
 
11
  ## Usage
12
  ```python
13
  from llmlingua import PromptCompressor
14
 
15
  compressor = PromptCompressor(
16
- model_name="microsoft/llmlingua-2-xlm-roberta-large-meetingbank",
17
- use_llmlingua2=True
18
- )
19
 
20
  original_prompt = """John: So, um, I've been thinking about the project, you know, and I believe we need to, uh, make some changes. I mean, we want the project to succeed, right? So, like, I think we should consider maybe revising the timeline.
21
  Sarah: I totally agree, John. I mean, we have to be realistic, you know. The timeline is, like, too tight. You know what I mean? We should definitely extend it.
22
  """
23
  results = compressor.compress_prompt_llmlingua2(
24
- original_prompt,
25
- rate=0.6,
26
- force_tokens=['\n', '.', '!', '?', ','],
27
- chunk_end_tokens=['.', '\n'],
28
- return_word_label=True,
29
- drop_consecutive=True
30
- )
31
 
32
  print(results.keys())
33
  print(f"Compressed prompt: {results['compressed_prompt']}")
@@ -50,5 +50,12 @@ for word, label in annotated_results[:10]:
50
 
51
  ## Citation
52
  ```
53
- {}
 
 
 
 
 
 
 
54
  ```
 
4
 
5
  # LLMLingua-2-Bert-base-Multilingual-Cased-MeetingBank
6
 
7
+ This model was introduced in the paper [**LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression** (Pan et al, 2024)](https://arxiv.org/abs/2403.12968). It is a [XLM-RoBERTa (large-sized model)](https://huggingface.co/FacebookAI/xlm-roberta-large) finetuned to perform token classification for task agnostic prompt compression. The probability $p_{preserve}$ of each token $x_i$ is used as the metric for compression. This model is trained on [an extractive text compression dataset(will public)]() constructed with the methodology proposed in the [**LLMLingua-2**](https://arxiv.org/abs/2403.12968), using training examples from [MeetingBank (Hu et al, 2023)](https://meetingbank.github.io/) as the seed data.
8
 
9
+ For more details, please check the home page of [LLMLingua-2](https://llmlingua.com/llmlingua2.html) and [LLMLingua Series](https://llmlingua.com/).
10
 
11
  ## Usage
12
  ```python
13
  from llmlingua import PromptCompressor
14
 
15
  compressor = PromptCompressor(
16
+ model_name="microsoft/llmlingua-2-xlm-roberta-large-meetingbank",
17
+ use_llmlingua2=True
18
+ )
19
 
20
  original_prompt = """John: So, um, I've been thinking about the project, you know, and I believe we need to, uh, make some changes. I mean, we want the project to succeed, right? So, like, I think we should consider maybe revising the timeline.
21
  Sarah: I totally agree, John. I mean, we have to be realistic, you know. The timeline is, like, too tight. You know what I mean? We should definitely extend it.
22
  """
23
  results = compressor.compress_prompt_llmlingua2(
24
+ original_prompt,
25
+ rate=0.6,
26
+ force_tokens=['\n', '.', '!', '?', ','],
27
+ chunk_end_tokens=['.', '\n'],
28
+ return_word_label=True,
29
+ drop_consecutive=True
30
+ )
31
 
32
  print(results.keys())
33
  print(f"Compressed prompt: {results['compressed_prompt']}")
 
50
 
51
  ## Citation
52
  ```
53
+ @article{wu2024llmlingua2,
54
+ title = "{LLML}ingua-2: Context-Aware Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression",
55
+ author = "Zhuoshi Pan and Qianhui Wu and Huiqiang Jiang and Menglin Xia and Xufang Luo and Jue Zhang and Qingwei Lin and Victor Ruhle and Yuqing Yang and Chin-Yew Lin and H. Vicky Zhao and Lili Qiu and Dongmei Zhang",
56
+ url = "https://arxiv.org/abs/2403.12968",
57
+ journal = "ArXiv preprint",
58
+ volume = "abs/2403.12968",
59
+ year = "2024",
60
+ }
61
  ```